SymmetricUncertainty<T>

Models & Types · AiDotNet.Preprocessing.FeatureSelection.Filter.InformationTheory

Symmetric Uncertainty for feature selection.

Inheritance: TransformerBase<T, Matrix<T>, Matrix<T>> → SymmetricUncertainty<T>

Implements: IFixedCountFeatureSelector<T>, ISupervisedFeatureSelector<T>

For Beginners

Regular Mutual Information can be hard to interpret because its scale depends on the data. Symmetric Uncertainty normalizes it to always be between 0 (no relationship) and 1 (perfect relationship). It treats both variables equally, so it doesn’t matter which is the feature and which is the target.

How It Works

Symmetric Uncertainty is a normalized version of Mutual Information that ranges from 0 to 1. It is symmetric, meaning SU(X,Y) = SU(Y,X), and corrects for the bias toward features with many values.

Constructors

Constructor Summary
SymmetricUncertainty(int nFeaturesToSelect, int nBins, int[] columnIndices)

Properties

Property Summary
Vector<T> SUScores { get; }
bool SupportsInverseTransform { get; }
int NBins { get; }
int NFeaturesToSelect { get; }
int[] SelectedIndices { get; }

Methods

Method Summary
Matrix<T> FitTransform(Matrix<T> data, Vector<T> target)
bool[] GetSupportMask()
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Matrix<T> data, Vector<T> target)